Appeal No. 2002-1197
Application 09/131,167
a lower full-scale value of the ADC, as recited in claims 5
and 29 (Br13). It is argued that column 4 of Koski does not
disclose or suggest threshold values, much less that the
threshold values are based in part on distortion through the ADC
or on an upper and lower full-scale value of an ADC (Br13-14).
The examiner states that Koski states (col. 1, lines 50-55)
that the reason for setting the threshold is to stop distortion
and therefore Koski is setting two thresholds to stop distortion
in the device (EA10).
We will not sustain the rejection. The thresholds in Koski,
columns 5 and 6, are for determining whether there is double talk
based on whether the power ratio (PR) of the far-end signal to
the near-end signal in the sub-channels are above or below the
threshold, taken in conjunction with the voice activity detector
(VAD) signal. Koski does not determine whether the output signal
is above or below a threshold, much less that the threshold
values relate to full-scale values of the ADC, as recited in
claims 5 and 29, or that at least one of the predetermined
threshold values is adjusted to account for ringing of the ADC
converter, as recited in claims 7 and 31. The fact that Koski
seeks to prevent distortion does not mean that it prevents
distortion in the way that is claimed. The rejection of
claims 5, 7, 29, and 31 is reversed.
- 13 -
Page: Previous 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Next
Last modified: November 3, 2007